312 REPORT 1851. 



tain this in a rudely approximate way, and subject also to this additional 

 source of error and uncertainty, that having no observation of the angle of 

 emergence of the wave of shock at the place of observation (and indeed very 

 seldom any apparent surface direction even given), and the apparent direction 

 of shock being almost always very far from identical with its real direction, 

 it becomes impossible to make the requisite trigonometrical allowance for the 

 change, from apparent to true direction, in modifying the observed velocity, 

 or rather the velocity as calculated, on the assumption that the wave travei'sed 

 a great circle of the earth's surface between the two points at which the times 

 of its transit had been observed ; still it is of value to compare such rude 

 approximations as we can yet command with the results we have already 

 experimentally obtained. 1 proceed to do so. 



In the Lisbon earthquake of 1761, the shock was recorded at the following 

 times and places : — 



H.M.S. Gosport off the Kock of Lisbon Cgiven as lat. ^i" 8' N. and long. 

 5° 10' W., apparently in error for 38° 4' N. lat., 10° 5' W. long.), shock felt 

 at a quarter of an hour before noon. 



At Corunna, felt at noon. 



At Cork, felt at 121^ 15"" p.m. 



At Santa Cruz, Barbary, at noon. 



As all these places are nearly in the same longitude, the clocks require no 

 correction for true time. The distances then are — 



Lisbon Rock to Corunna . . S^O statute miles. 



Lisbon Rock to Cork 903 „ 



Lisbon Rock to Santa Cruz . 556 „ 



or in time — 



340 statute miles in 15" = 22'66 miles per minute. 



903 15"^ = 60-0 



556 15"= 37-06 



The greatest of these transit rates occurs in a range under the Bay of 

 Biscay, consisting probably to a great extent of hard consistent slates or other 

 still more elastic rocks. 



M. Peri'ey, in his Memoir on Earthquakes in the Antilles (Mem. Acad, de 

 Dijon, 1845-46, p. 367), in recording the well-known shocks of Pointe a Pitre 

 in Guadaloupe, of February 8, 1843, says, •' M. Itier is of opinion that the 

 shock in a direction from N.W. to S.E. took thirteen minutes to travel from 

 Guadaloupe to Cayenne in South America. This distance is nearly 14°=840 

 geographical = 973 statute miles, which is at the rate of 74*84 statute miles 

 per minute." 



M. Perrey attaches the greatest possible doubt to the correctness of M. 

 Itier's records and observations in this matter. The superficial material 

 forming the intermediate range is probably principally of soft diluvial or 

 detrital material ; but if the shock were transferred laterally, as is probable, at 

 a great depth, and was felt transversal to its principal line of transit at 

 Cayenne, it is likely to have had its range in very hard and elastic rock, 

 though of what precise character there is no information. 



The two following East Indian earthquakes are of much interest in refer- 

 ence to the point before us. The skeleton map of India (PI. XVII.) I have 

 prepared by reduction from Allen's great map, and approximately divided its 

 surface into coloured geological divisions, having reference, not so much to 

 recognised geological formations, as to the classification of formations to- 

 gether under one tint, that possess something like a general equality of pro- 

 bable transmissive power for earthquake-waves, so that the whole surface 

 has been massed under six divisions, viz.— 1 , crystalline, schistose, or granitoid 



